HL.mean {asbio}

Hodges-Lehman estimator of location

Description

Calculates the Hodges-Lehman estimate of location –which is consistent for the true pseudomedian– using Walsh averages (Hollander and Wolfe 1999, pgs. 51-55). If requested, the function also provides confidence intervals for the true pseudomedian. In a symmetric distribution the mean, median, and pseudomedian will be identical.

Usage

HL.mean(x, conf = NULL, method = "exact")

Arguments

x

A vector of quantitative data.

conf

A proportion specifying 1 - P(type I error).

method

method for confidence interval calculation. One of "approx", which uses a normal approximation, or "exact", which uses the Wilcoxon sign-rank quantile function (see Hollander and Wolfe 1999).

Author(s)

Ken Aho

References

Hollander, M., and Wolfe, D. A (1999) Nonparametric Statistical Methods. New York: John Wiley & Sons.

See Also

H.mean, G.mean

Examples

# Hamilton depression scale (Hollander and Wolfe 1999)
x<-c( -0.952, 0.147, -1.022, -0.430, -0.620, -0.590, -0.490,  0.080, -0.010)
HL.mean(x, conf = .96)